The power of possibility: causal learning, counterfactual reasoning, and pretend play

We argue for a theoretical link between the development of an extended period of immaturity in human evolution and the emergence of powerful and wide-ranging causal learning mechanisms, specifically the use of causal models and Bayesian learning. We suggest that exploratory childhood learning, childhood play in particular, and causal cognition are closely connected. We report an empirical study demonstrating one such connection--a link between pretend play and counterfactual causal reasoning. Preschool children given new information about a causal system made very similar inferences both when they considered counterfactuals about the system and when they engaged in pretend play about it. Counterfactual cognition and causally coherent pretence were also significantly correlated even when age, general cognitive development and executive function were controlled for. These findings link a distinctive human form of childhood play and an equally distinctive human form of causal inference. We speculate that, during human evolution, computations that were initially reserved for solving particularly important ecological problems came to be used much more widely and extensively during the long period of protected immaturity.     

The nucleotide sequences recognized by endonucleases AvaI and AvaII from Anabaena variabilis.

Determination of the 5'-terminal sequences flanking all the individual cleavage sites for endonuclease AvaI in bacteriophage-lambda DNA has shown that this enzyme recognizes the hexanucleotide sequences: (Formula: see text), This sequence is cut as shown by the arrows to give single-stranded 5'-tetranucleotide protrusions (cohesive ends). Endonucleases SmaI, XhoI and XmaI recognize different symmetrical subsets of this sequence and provide independent evidence for the occurrence of these subsets at particular endonuclease-AvaI cleavage sites in the bacteriophage-lambda genome. Further evidence for this structure came from the demonstration that DNA fragments generated by endonuclease AvaI can be ligated to form a discrete set of larger molecules and from nearest-neighbor analysis which showed that cytosine residues occurred at the 3'-side of cleavage points. The observation that endonuclease AvaII recognized a subset of the sites recognized by AsuI [Hughes, Bruce & Murray (1979) Biochem. J. 185, 59-63[ led to the deduction that AvaII recognize the pentanucleotide sequence: (Formula: see text), and breaks internucleotide bonds at the positions indicated by the arrows.     

Impulsivity, compulsivity, and top-down cognitive control

Impulsivity is the tendency to act prematurely without foresight. Behavioral and neurobiological analysis of this construct, with evidence from both animal and human studies, defines several dissociable forms depending on distinct cortico-striatal substrates. One form of impulsivity depends on the temporal discounting of reward, another on motor or response disinhibition. Impulsivity is commonly associated with addiction to drugs from different pharmacological classes, but its causal role in human addiction is unclear. We characterize in neurobehavioral and neurochemical terms a rodent model of impulsivity based on premature responding in an attentional task. Evidence is surveyed that high impulsivity on this task precedes the escalation subsequently of cocaine self-administration behavior, and also a tendency toward compulsive cocaine-seeking and to relapse. These results indicate that the vulnerability to stimulant addiction may depend on an impulsivity endophenotype. Implications of these findings for the etiology, development, and treatment of drug addiction are considered.     

Information integration: its relevance to brain function and consciousness

A proper understanding of cognitive functions cannot be achieved without an understanding of consciousness, both at the empirical and at the theoretical level. This paper argues that consciousness has to do with a system's capacity for information integration. In this approach, every causal mechanism capable of choosing among alternatives generates information, and information is integrated to the extent that it is generated by a system above and beyond its parts. The set of integrated informational relationships generated by a complex of mechanisms--its quale--specify both the quantity and the quality of experience. As argued below, depending on the causal structure of a system, information integration can reach a maximum value at a particular spatial and temporal grain size. It is also argued that changes in information integration reflect a system's ability to match the causal structure of the world, both on the input and the output side. After a brief review suggesting that this approach is consistent with several experimental and clinical observations, the paper concludes with some prospective remarks about the relevance of understanding information integration for analyzing cognitive function, both normal and pathological.     

Causal measures of structure and plasticity in simulated and living neural networks

A major goal of neuroscience is to understand the relationship between neural structures and their function. Recording of neural activity with arrays of electrodes is a primary tool employed toward this goal. However, the relationships among the neural activity recorded by these arrays are often highly complex making it problematic to accurately quantify a network's structural information and then relate that structure to its function. Current statistical methods including cross correlation and coherence have achieved only modest success in characterizing the structural connectivity. Over the last decade an alternative technique known as Granger causality is emerging within neuroscience. This technique, borrowed from the field of economics, provides a strong mathematical foundation based on linear auto-regression to detect and quantify "causal" relationships among different time series. This paper presents a combination of three Granger based analytical methods that can quickly provide a relatively complete representation of the causal structure within a neural network. These are a simple pairwise Granger causality metric, a conditional metric, and a little known computationally inexpensive subtractive conditional method. Each causal metric is first described and evaluated in a series of biologically plausible neural simulations. We then demonstrate how Granger causality can detect and quantify changes in the strength of those relationships during plasticity using 60 channel spike train data from an in vitro cortical network measured on a microelectrode array. We show that these metrics can not only detect the presence of causal relationships, they also provide crucial information about the strength and direction of that relationship, particularly when that relationship maybe changing during plasticity. Although we focus on the analysis of multichannel spike train data the metrics we describe are applicable to any stationary time series in which causal relationships among multiple measures is desired. These techniques can be especially useful when the interactions among those measures are highly complex, difficult to untangle, and maybe changing over time.     

Clinical trials of fatty acid treatment in ADHD, dyslexia, dyspraxia and the autistic spectrum

Considerable clinical and experimental evidence now supports the idea that deficiencies or imbalances in certain highly unsaturated fatty acids may contribute to a range of common developmental disorders including ADHD, dyslexia, dyspraxia and autistic spectrum disorders (ASD). Definitive evidence of a causal contribution, however, can only come from intervention studies in the form of randomised, double-blind, placebo-controlled trials. Published studies of this kind are still fairly few in number, and mainly involve the diagnostic categories of ADHD and dyslexia, although other trials involving individuals with dyspraxia or ASD are in progress. The main findings to date from such studies are reviewed and evaluated here with the primary aim of guiding future research, although given that fatty acid supplementation for these conditions is already being adopted in many quarters, it is hoped that some of the information provided may also help to inform clinical practice.     

Natural selection. VII. History and interpretation of kin selection theory

Kin selection theory is a kind of causal analysis. The initial form of kin selection ascribed cause to costs, benefits and genetic relatedness. The theory then slowly developed a deeper and more sophisticated approach to partitioning the causes of social evolution. Controversy followed because causal analysis inevitably attracts opposing views. It is always possible to separate total effects into different component causes. Alternative causal schemes emphasize different aspects of a problem, reflecting the distinct goals, interests and biases of different perspectives. For example, group selection is a particular causal scheme with certain advantages and significant limitations. Ultimately, to use kin selection theory to analyse natural patterns and to understand the history of debates over different approaches, one must follow the underlying history of causal analysis. This article describes the history of kin selection theory, with emphasis on how the causal perspective improved through the study of key patterns of natural history, such as dispersal and sex ratio, and through a unified approach to demographic and social processes. Independent historical developments in the multivariate analysis of quantitative traits merged with the causal analysis of social evolution by kin selection.     

Hawks, doves, and mixed-symmetry games

The hawk-dove game has proved to be an important tool for understanding the role of aggression in social interactions. Here, the game is presented in a more general form (GHD) to facilitate analyses of interactions between individuals that may differ in "size", where size is interpreted as a surrogate for resource holding power. Three different situations are considered, based on the availability and use of information that interacting individuals have about their sizes: the classical symmetric case, in which no information about sizes is used, the asymmetric case, in which the individuals know their relative sizes and thus their chances of prevailing in combat, and a mixed-symmetry case, in which each individual only knows its own size (or only knows its opponent's size). I describe and use some recently developed methods for multitype games-evolutionary games involving two or more categories of players. With these methods and others, the evolutionarily stable strategies (ESSs) that emerge for the three different cases are identified and compared. A proof of the form and uniqueness of the ESS for the mixed-symmetry case is presented. In this situation, one size category at most can play a mixed strategy; larger individuals are aggressive and smaller individuals are not. As the number of size categories approaches infinity and the size distribution becomes continuous, there is a threshold size, above which all individuals are aggressive, and below which they are not.     

Evolution, Causal Biology, and Classification

Brundin, L. (Section of Entomology, Swedish Museum of Natural History, Stockholm, Sweden.) Evolution, causal biology, and classification. Zool. Scripta 1 (3–4): 107–120, 1972.–The pros and cons of different approaches to classification are discussed against the background of a critical survey of the leading principles of evolution and the methodological advices furnished by the nature of the evolutionary process for the attainment of an adequate reference system of causal biology. It is shown that such a system has to be a reconstruction of nature's own hierarchy. The reasons for the present disagreement between the Hennig school and the Simpson-Mayr school are discussed at some length, and it is stressed that the classificatory dilemma of the latter school is an unavoidable consequence of its quantitative approach and hence self-inflicted.     

Bifungites, trace fossils from Devonian-Mississippian rocks of Pennsylvania and Montana,U.S.A

The trace fossil Bifungites Desio 1940, first recognized in Late Devonian rocks of Libya and later in Algeria, is common in Late Devonian rocks of Montana, Pennsylvania, Ohio, and Michigan, and Early Mississippian rocks of Pennsylvania. Its manifestations are given for seven stratigraphic units from forty localities in Pennsylvania, Montana, and Michigan.Bifungites commonly occur on bedding planes and consist of a horizontal shaft with doubly terminating arrow- or dumb-bell-shaped projections. Less apparent are paired vertical cylindrical sediment-filled tubes connected at their base to the bi-arrow or dumb-bell shaft. The combination represents the mud or silt casting of an inverted pi-shaped, μ, infaunal tubular burrow (domichnia). Siltstone slabs have been transversely sectioned to reveal this π-shaped, double-arrow burrow pattern with undisturbed stratification between the vertical tubes.Three new ichnospecies are proposed based on markedly different size and shape of the fossil traces in three stratigraphic zones. A Late Devonian Girard Shale type is the smallest with wide, short arrows on a narrow shaft whose median overall length is 12.5 mm. The latest Devonian Sappington type has well-formed, shortly barbed arrow terminations and median length of 28 mm. An Early Mississippian Meadville type is the largest with median length of 36 mm and prominently barbed broad arrows.The Bifungites organism is unrecognized but it was probably a sedentary, soft-bodied, infaunal suspension-feeding animal inhabiting shallow nearshore marine and brackish water environments. The identified biota directly or indirectly associated with Bifungites is extremely limited to a few brachiopods, clams, and other trace fossils.     

Multisensory Causal Inference in the Brain

At any given moment, our brain processes multiple inputs from its different sensory modalities (vision, hearing, touch, etc.). In deciphering this array of sensory information, the brain has to solve two problems: (1) which of the inputs originate from the same object and should be integrated and (2) for the sensations originating from the same object, how best to integrate them. Recent behavioural studies suggest that the human brain solves these problems using optimal probabilistic inference, known as Bayesian causal inference. However, how and where the underlying computations are carried out in the brain have remained unknown. By combining neuroimaging-based decoding techniques and computational modelling of behavioural data, a new study now sheds light on how multisensory causal inference maps onto specific brain areas. The results suggest that the complexity of neural computations increases along the visual hierarchy and link specific components of the causal inference process with specific visual and parietal regions.     

Increasing power of groupwise association test with likelihood ratio test

Sequencing studies have been discovering a numerous number of rare variants, allowing the identification of the effects of rare variants on disease susceptibility. As a method to increase the statistical power of studies on rare variants, several groupwise association tests that group rare variants in genes and detect associations between genes and diseases have been proposed. One major challenge in these methods is to determine which variants are causal in a group, and to overcome this challenge, previous methods used prior information that specifies how likely each variant is causal. Another source of information that can be used to determine causal variants is the observed data because case individuals are likely to have more causal variants than control individuals. In this article, we introduce a likelihood ratio test (LRT) that uses both data and prior information to infer which variants are causal and uses this finding to determine whether a group of variants is involved in a disease. We demonstrate through simulations that LRT achieves higher power than previous methods. We also evaluate our method on mutation screening data of the susceptibility gene for ataxia telangiectasia, and show that LRT can detect an association in real data. To increase the computational speed of our method, we show how we can decompose the computation of LRT, and propose an efficient permutation test. With this optimization, we can efficiently compute an LRT statistic and its significance at a genome-wide level. The software for our method is publicly available at http://genetics.cs.ucla.edu/rarevariants .     

Estimating Temporal Causal Interaction between Spike Trains with Permutation and Transfer Entropy

Estimating the causal interaction between neurons is very important for better understanding the functional connectivity in neuronal networks. We propose a method called normalized permutation transfer entropy (NPTE) to evaluate the temporal causal interaction between spike trains, which quantifies the fraction of ordinal information in a neuron that has presented in another one. The performance of this method is evaluated with the spike trains generated by an Izhikevich’s neuronal model. Results show that the NPTE method can effectively estimate the causal interaction between two neurons without influence of data length. Considering both the precision of time delay estimated and the robustness of information flow estimated against neuronal firing rate, the NPTE method is superior to other information theoretic method including normalized transfer entropy, symbolic transfer entropy and permutation conditional mutual information. To test the performance of NPTE on analyzing simulated biophysically realistic synapses, an Izhikevich’s cortical network that based on the neuronal model is employed. It is found that the NPTE method is able to characterize mutual interactions and identify spurious causality in a network of three neurons exactly. We conclude that the proposed method can obtain more reliable comparison of interactions between different pairs of neurons and is a promising tool to uncover more details on the neural coding.     

Identifying interactions in the time and frequency domains in local and global networks - A Granger Causality Approach

Background  

Reverse-engineering approaches such as Bayesian network inference, ordinary differential equations (ODEs) and information theory are widely applied to deriving causal relationships among different elements such as genes, proteins, metabolites, neurons, brain areas and so on, based upon multi-dimensional spatial and temporal data. There are several well-established reverse-engineering approaches to explore causal relationships in a dynamic network, such as ordinary differential equations (ODE), Bayesian networks, information theory and Granger Causality.     

Moral Darwinism: Ethical evidence for the descent of man

Could an ethical theory ever play a substantial evidential role in a scientific argument for an empirical hypothesis? InThe Descent of Man, Darwin includes an extended discussion of the nature of human morality, and the ethical theory which he sketches is not simply developed as an interesting ramification of his theory of evolution, but is used as a key part of his evidence for human descent from animal ancestors. Darwin must rebut the argument that, because of our moral nature, humans are essentially different in kind from other animals and so had to have had a different origin. I trace his causal story of how the moral sense could develop out of social instincts by evolutionary mechanisms of group selection, and show that the form of Utilitarianism he proposes involves a radical reduction of the standard of value to the concept of biological fitness. I argue that this causal analysis, although a weakness from a normative standpoint, is a strength when judged for its intended purpose as part of an evidential argument to confirm the hypothesis of human descent.     

Gravity, light and plant form

Plants have evolved highly sensitive and selective mechanisms that detect and respond to various aspects of their environment. As a plant develops, it integrates the environmental information perceived by all of its sensory systems and adapts its growth to the prevailing environmental conditions. Light is of critical importance because plants depend on it for energy and, thus, survival. The quantity, quality and direction of light are perceived by several different photosensory systems that together regulate nearly all stages of plant development, presumably in order to maintain photosynthetic efficiency. Gravity provides an almost constant stimulus that is the source of critical spatial information about its surroundings and provides important cues for orientating plant growth. Gravity plays a particularly important role during the early stages of seedling growth by stimulating a negative gravitropic response in the primary shoot that orientates it towards the source of light, and a positive gravitropic response in the primary root that causes it to grow down into the soil, providing support and nutrient acquisition. Gravity also influences plant form during later stages of development through its effect on lateral organs and supporting structures. Thus, the final form of a plant depends on the cumulative effects of light, gravity and other environmental sensory inputs on endogenous developmental programs. This article is focused on developmental interactions modulated by light and gravity.     

Estimating the directed information to infer causal relationships in ensemble neural spike train recordings

Advances in recording technologies have given neuroscience researchers access to large amounts of data, in particular, simultaneous, individual recordings of large groups of neurons in different parts of the brain. A variety of quantitative techniques have been utilized to analyze the spiking activities of the neurons to elucidate the functional connectivity of the recorded neurons. In the past, researchers have used correlative measures. More recently, to better capture the dynamic, complex relationships present in the data, neuroscientists have employed causal measures—most of which are variants of Granger causality—with limited success. This paper motivates the directed information, an information and control theoretic concept, as a modality-independent embodiment of Granger’s original notion of causality. Key properties include: (a) it is nonzero if and only if one process causally influences another, and (b) its specific value can be interpreted as the strength of a causal relationship. We next describe how the causally conditioned directed information between two processes given knowledge of others provides a network version of causality: it is nonzero if and only if, in the presence of the present and past of other processes, one process causally influences another. This notion is shown to be able to differentiate between true direct causal influences, common inputs, and cascade effects in more two processes. We next describe a procedure to estimate the directed information on neural spike trains using point process generalized linear models, maximum likelihood estimation and information-theoretic model order selection. We demonstrate that on a simulated network of neurons, it (a) correctly identifies all pairwise causal relationships and (b) correctly identifies network causal relationships. This procedure is then used to analyze ensemble spike train recordings in primary motor cortex of an awake monkey while performing target reaching tasks, uncovering causal relationships whose directionality are consistent with predictions made from the wave propagation of simultaneously recorded local field potentials.